SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Low bone formation is considered to be the main feature in osteoporosis associated with cholestatic and end-stage liver diseases, although the consequences of retained substances in chronic cholestasis on bone cells have scarcely been studied. Therefore, we analyzed the effects of bilirubin and serum from jaundiced patients on viability, differentiation, mineralization, and gene expression in the cells involved in bone formation. The experiments were performed in human primary osteoblasts and SAOS-2 human osteosarcoma cells. Unconjugated bilirubin or serum from jaundiced patients resulted in a dose-dependent decrease in osteoblast viability. Concentrations of bilirubin or jaundiced serum without effects on cell survival significantly diminished osteoblast differentiation. Mineralization was significantly reduced by exposure to 50 μM bilirubin at all time points (from −32% to −55%) and jaundiced sera resulted in a significant decrease on cell mineralization as well. Furthermore, bilirubin down-regulated RUNX2 (runt-related transcription factor 2) gene expression, a basic osteogenic factor involved in osteoblast differentiation, and serum from jaundiced patients significantly up-regulated the RANKL/OPG (receptor activator of nuclear factor-κB ligand/osteoprotegerin) gene expression ratio, a system closely involved in osteoblast-induced osteoclastogenesis. Conclusion: Besides decreased cell viability, unconjugated bilirubin and serum from jaundiced patients led to defective consequences on osteoblasts. Moreover, jaundiced serum up-regulates the system involved in osteoblast-induced osteoclastogenesis. These results support the deleterious consequences of increased bilirubin in advanced chronic cholestasis and in end-stage liver diseases, resulting in disturbed bone formation related to osteoblast dysfunction. (HEPATOLOGY 2011)

The pathogenesis of osteoporosis in patients with chronic cholestasis and in those with end-stage liver disease is not well understood.1, 2 Thus, both low bone formation3 and increased resorption have been described.4 Although a number of studies have been performed to elucidate the risk factors for osteoporosis and metabolic bone disease in patients with these conditions, few pathophysiological assessments have been carried out to delineate the intrinsic factors participating in the development of bone disease. In this respect, it has been proposed that osteoporosis may result from the damaging effect of retained substances such as bilirubin and bile acids on osteoblasts, which are the cells involved in bone formation. One study demonstrated that unconjugated bilirubin has a detrimental effect on the viability of cultured human osteoblasts with no effect on bile acids.5 However, we recently observed that lithocholic acid, a monohydroxylated secondary bile acid, synthesized mainly from the intestinal bacterial 7-dehydroxylation of chenodeoxycholic acid, has deleterious effects on human osteoblasts, not only in relation to their viability, but also regarding the potential damaging effects of lithocholic acid on the vitamin D pathways, through the vitamin D receptor.6

Despite the clear-cut association between low bone mass and jaundice in patients with chronic cholestatic diseases,7-9 and the experimental evidence of skeletal fragility in bile duct–ligated rats,10 the influence of bilirubin on osteoporosis in patients with liver disease has been questioned because of some contradictory results concerning low bone mass and Gilbert's syndrome, a mild clinical condition characterized by increased circulating levels of unconjugated bilirubin.11, 12 The effects of bilirubin on osteoblasts have only been assessed in one study which was mainly focused on the consequences on cell viability.5 However, no other effects have been explored including cell differentiation and mineralization and the regulation of some osteoblast-related genes, particularly those from the osteoprotegerin (OPG)/receptor activator of nuclear factor-κB ligand (RANKL) system, which are the key regulators of osteoblast-induced osteoclastogenesis. Therefore, in this study, we evaluated the consequences of high bilirubin concentrations on cell viability, differentiation, mineralization, and gene expression in osteoblasts.

Patients and Methods

  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Materials.

Dulbecco's modified Eagle medium (DMEM), Ham's formula-12 nutrient mixture (HAM F-12), Hank's balanced salt solution (HBSS), fetal bovine serum (FBS), L-glutamine, and trypsin were purchased from Invitrogen (Grand Island, NY); insulin–transferrin–selenium, dihydroxyvitamin D3 (vitamin D), bilirubin, ascorbic acid, α-naphthylphosphate acid, and fast blue were from Sigma Chemical Co. (St. Louis, MO); penicillin–streptomycin was obtained from LabClinics (Barcelona, Spain).

Experimental Bilirubin Solution Preparation.

Bilirubin (Sigma) stock solution of 1600 μM was prepared just before use by dissolving it into 10 mL 0.1 N NaOH under dim light, as described.13, 14 The bilirubin solution was passed through a sterile filter (0.22 μm pore size) and adjusted to pH 7.2-7.4 with 0.1 N HCl if necessary. The bilirubin stock solution was added to a final concentration of 10 to 1000 μM in the culture medium. The cell cultures were kept in dark conditions to prevent light degradation of the bilirubin. Control cells were treated with vehicle (NaOH 0.1 N).

Samples from Patients and Healthy Subjects.

A 20-mL blood sample was obtained from 15 jaundiced patients (seven with primary biliary cirrhosis, three with cholestatic acute hepatitis of unknown etiology, and five with a cholestatic form of severe alcoholic hepatitis), 30 patients with primary biliary cirrhosis and normal bilirubin levels (50% with densitometric osteopenia, and 50% with normal bone densitometry), and 30 healthy subjects who were matched by age and sex. No patients were under treatment with glucocorticoids, bisphosphonates, or calcitonin, and all had normal serum calcium concentrations. The mean total serum bilirubin concentration measured by standard colorimetric assay in jaundiced patients was 330 μM (19.3 mg/dL). Sera was separated from the blood samples and stored at −80°C until the assays were performed. Samples were mixed in equal volume condition of conventional culture medium. To avoid photodegradation, all bilirubin-containing serum samples were prepared in the dark, and all cell culture plates were wrapped in aluminum foil to avoid light exposure. The experiments were carried out with pooled samples. The protocol was reviewed and approved by the Hospital Clínic, and all subjects gave informed consent.

Cell Cultures.

The experiments were performed with human primary osteoblasts and SAOS-2 human osteosarcoma cells. Human primary osteoblasts were taken from trabecular bone specimens using a modification of the procedure of Robey and Termine.15 Bone pieces were obtained from subjects without features of metabolic bone disease who were undergoing hip replacement for osteoarthritis, following the procedures approved by the Hospital Clínic Ethics Committee. Trabecular bone pieces, processed according to previously described protocol,16 were grown in DMEM/HAM F-12 (1:1) medium, supplemented with 10% heat-inactivated FBS, 100 U/mL penicillin–streptomycin, and 10 μg/mL ascorbic acid. Cells were characterized as osteoblast phenotypes by determination of alkaline phosphatase activity4 and osteocalcin messenger RNA (mRNA) expression by histochemistry and reverse-transcription polymerase chain reaction (PCR), respectively. Only cells in the first passage were used in the experiments.

Cell mineralization was carried out with SAOS-2 human osteosarcoma cells from ATCC (American Type Culture Collection, Manassas, VA), which were cultured in DMEM supplemented with 10% FBS. The cells were incubated at 37°C in a humidified atmosphere of 95% air and 5% CO2, and the medium was changed twice a week. This cell line was used because it is the common model for mineralization assessment for calcium deposition capacity in osteogenic conditions.17-19

Characterization of Primary Osteoblasts.

Cells were characterized as human osteoblastic cells by determination of the alkaline phosphatase activity, as measured by histochemical technique. Cells were placed on 35-mm coverslips at a density of 4 × 105 cells/mL and incubated for 72 hours at 37°C with DMEM/HAM F-12 containing 10 μg/mL ascorbic acid and 1,25-dihydroxycholecalciferol at 10−7 M to stimulate the alkaline phosphatase synthesis, and the enzyme activity was assessed in cells grown to confluence. Cells were rinsed twice with HBSS and fixed in cold 95% ethanol, then were incubated at room temperature with a solution of α-naphthylphosphate and 0.1% Tris-buffered HCl at pH 10 containing 0.1% fast blue RR. Stained cells were identified by optical microscopy.

Human osteoblastic cells were also characterized by qualitative osteocalcin PCR. Briefly, cell cultures were carried out as previously described, and RNA extraction and reverse-transcription were performed using Trizol Reagent and Ready-To-Go First Strand kit (see section on RNA isolation and gene expression by real-time PCR). The PCR reaction was performed as previously described, in a 50 μL reaction mixture containing 5 μL complementary DNA, 20 mM Tris-HCl, 50 mM MgCl2, 200 μM of each deoxynucleotide triphosphate, 0.3 mM of each specific primer (sense: 5′-TCA CAC TCC TCG CCC TAT T-3′ and antisense: 5′-CGA TGT GGT CAG CCA ACT-3′), and 0.03 U/μL Taq DNA polymerase (GibcoBRL, Grand Island, NY). PCR of β-actin was used as an endogenous control. The expected sizes of the PCR products of osteocalcin and β-actin were 246 and 285 base pairs, respectively.

Cell Viability Assay.

A pool of primary osteoblastic cells from 10 donors were plated in 24-well tissue plates and were incubated in DMEM/HAM F-12 (1:1) medium, supplemented with 10% of FBS and 10 μg/mL of ascorbic acid. In order to synchronize after reaching osteoblast subconfluence, culture medium was replaced with DMEM/HAM F-12 containing 100 μg/mL of ascorbic acid and incubated for 24 hours. Cells were subsequently incubated for 24, 48, and 72 hours with different concentrations of unconjugated bilirubin (10, 50, 100, and 1000 μM) or pooled samples from cholestatic patients with normal and high bilirubin levels, and pooled samples from healthy controls, in DMEM/HAM F-12 medium with 10 μg/mL ascorbic acid. Cell viability was measured in duplicate using a colorimetric assay based on the cleavage of the tetrazolium salt WST-1 by mitochondrial dehydrogenase in viable cells (Cell Proliferation Reagent WST-1; Roche, Basel, Switzerland). The absorbance was read at 450 nm wavelength with an enzyme-linked immunosorbent assay reader.

Cell Differentiation.

Osteoblast differentiation was measured by the determination of alkaline phosphatase activity. Briefly, primary osteoblasts from three subjects were plated in 12-well tissue plates and incubated in supplemented medium. After synchronization, cells were incubated for 24, 48, and 72 hours in 10 μg/mL of ascorbic acid with different concentrations of unconjugated bilirubin (10, 50, 100, and 1000 μM) or pooled samples from patients with normal and high bilirubin levels, and samples from healthy controls. Then, cells were washed with phosphate-buffered saline and lysed with a lysis buffer (CelLytic M; Sigma Aldrich). Cell extracts were incubated with 2 mg/mL of p-nitrophenylphosphate (pNPP) in a 0.05 M glycine buffer containing 0.5 mM MgCl2 (pH 10.5) at 37°C for 30 minutes. The reaction was stopped by the addition of 0.4 N NaOH to the reaction mixture, and the alkaline phosphatase activity was quantified by absorbance at 405 nm. Total protein content was determined with Bradford's method in aliquots of the same samples with the Quick Start Bradford Protein Assay (Bio-Rad Laboratories, Madrid, Spain). Alkaline phosphatase levels were expressed by the total protein content. All measurements were performed in triplicate. A primary human fibroblast cell line was used as a negative control of alkaline phosphatase activity.

Osteoblast Mineralization.

The procedure was carried out by measuring the mineralized nodule formation using the alizarin red staining. Briefly, SAOS-2 cells were plated onto 12-well culture plates at a density of 105 cells/well in DMEM at 10% FBS with 50 μg/mL L-ascorbic acid and 10 mM β-glycerophosphate. After confluence, cells were treated with either different unconjugated bilirubin concentrations (10 and 50 μM) or pooled samples from patients with normal and high bilirubin levels and samples from healthy subjects, for 7, 14, 21, and 28 days. The culture media was changed every 3 days and with the same concentrations of the bilirubin and plasma samples. After 1 week, cells were washed with phosphate-buffered saline, fixed with 10% formaldehyde, and incubated at room temperature for 15 minutes. After cells were rinsed three times (5-10 minutes each) with an excess of distilled water, 1 mL/well of alizarin red stain solution (1% pH: 4.2) was added, and cells were incubated at room temperature for at least 20 minutes. Differentiated cells containing mineral deposits were brightly stained in red.

To quantify the alizarin red staining, 400 μL 10% acetic acid was added to each well of a 24-well plate and incubated for 30 minutes while shaking. Cells were then gently scraped and transferred to a microcentrifuge tube. After vigorously vortexing for 30 seconds, the cellular suspension was heated to 85°C for 10 minutes and then immediately kept on ice for 5 minutes. Finally, the slurry was centrifuged at 20,000g for 15 minutes, and 400 μL of the supernatant was removed and transferred to a new microcentrifuge tube. To neutralize the pH, ∼150 μL 10% ammonium hydroxide was added, and the absorbance solution was read at 405 nm wavelength. The calcium concentration was calculated according to a standard curve and normalized by the total protein content. All measurements were performed in triplicate. One mole alizarin red-S selectively binds approximately 2 mol of calcium.

RNA Isolation and Gene Expression by Real-Time PCR.

The osteoblast differentiation markers such as osteocalcin and runt-related transcription factor 2 (RUNX2) were examined, in addition to the expression of other genes expressed in osteoblasts such as osteoprotegerin (OPG) and osteoclast receptor activator of nuclear factor-κB ligand (RANKL). A pool of primary osteoblastic cells from 10 donors was plated in six-well tissue plates and was incubated according to the conditions indicated above during 24 hours. Total cellular RNA was extracted from osteoblasts grown in culture using an acid guanidinium–phenol–chloroform method (Trizol reagent; Invitrogen) according to the manufacturer's protocols. RNA content was determined using A260/A280 (absorbance at 260 and 280 nm) spectrophotometry. RNA was reverse-transcribed from 2 μg of total RNA from each sample using a complementary DNA synthesis kit (Ready-to-Go First Strand kit; Amersham Biosciences, Freiburg, Germany) and deoxythymidine oligomer as a primer.

Designed human TaqMan assays (Applied Biosystems, Foster City, CA) were used to quantify gene expression of osteocalcin (BGLAP), osteoprotegerin (TNFRSF11B), RANKL (TNFSF11), and Cbfa1 (RUNX2). Quantitative PCRs were carried out using ABI-Prism 7900 HT Fast Real-Time PCR System and a TaqMan 5′-nuclease probe method (Applied Biosystems). Results were expressed as relative expression of each gene (versus β-actin gene expression), using arbitrary units according to the comparative CT (threshold cycle) method.20 All real-time PCR reactions for each sample were performed in triplicate. The primers used are listed in Table 1.

Table 1. List of Primers Used in Real-Time PCR Reactions for Each Gene
GeneIdentification NumberTaqMan MGB Probe Sequences
  1. MGB, major groove binder.

OPGHs00171068 m15′-CGCGCTCGTGTTTCTGGACATCTCC-3′
RANKLHs00243522 m15′-CCGACATCCCATCTGGTTCCCATAA-3′
OsteocalcinHs00609452 g15′-CAAAGGTGCAGCCTTTGTGTCCAAG-3′
RUNX2Hs00231692 m15′-TCGGGAACCCAGAAGGCACAGACAG-3′
β-ActinHs99999903 m15′-TCGCCTTTGCCGATCCGCCGCCCGT-3′

Statistics.

Data are expressed as mean ± standard deviation (SD). All analyses were performed with the SPSS version 14.00 statistical package (SPSS Inc., Chicago, IL). Significant differences between any two groups were determined by Student t test or Mann-Whitney U test. When multiple groups were compared, analysis of variance was used, followed by a Tukey's multiple contrast test, where applicable. A P value ≤0.05 was considered significant.

Results

  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Nonpassage human primary osteoblasts, after synchronization, displayed the characteristic pattern of gene expression and protein production of osteoblastic differentiation markers such as osteocalcin gene expression and alkaline phosphatase activity (data not shown).

Effect of Bilirubin Concentration and Samples from Patients and Healthy Subjects on Osteoblast Viability.

Increasing concentrations of unconjugated bilirubin in the culture media resulted in a progressive decrease in cell viability, which was observed particularly at concentrations higher than 100 μM at 48 hours and higher than 50 μM at 72 hours (Table 2). The cell viability decrease was 36% and 56%, at 50 and 100 μM, respectively, compared with nontreated cells. Moreover, the presence of bilirubin (10 μM) resulted in significantly better cell viability compared with no bilirubin in the experiments performed at 48 hours and in the plates without FBS. These effects on cell survival were partially prevented by the presence of 10% FBS, because the detrimental effect of bilirubin at 50 and 100 μM was completely abolished in the experiments with FBS. Actually, in these latter experiments, the decreased cell viability was only observed with bilirubin at 1000 μM (Table 2).

Table 2. Effect of Bilirubin on Osteoblast Viability
Time (hours)Bilirubin μM (mg/dL)P Value
0 (0)10 (0.6)50 (2.9)100 (5.8)1000 (58.5)
  • *

    P < 0.01 versus 0 μM;

  • **

    P < 0.01 versus 10 μM;

  • ***

    P < 0.01 versus 0 and 10 μM.

  • P Value: Multiple groups comparation. Significant differences are shown in bold/italic. Viability of primary human osteoblast cells of three pools from 10 donors each, incubated with increasing concentrations of bilirubin (10, 50, 10, and 1000 μM) (0, 0.6, 2.9, 5.8, and 58.5 mg/dL, respectively) for 24, 48, and 72 hours. Cell viability is expressed as percentage of absorbance versus nontreated cells (control). Culture experiments were performed in triplicate for each pool of human osteoblast cells and treatments.

No FBS
24100136.3 ± 12.9134.7 ± 14.2127.5 ± 24.720.6 ± 8.8***0.000
48100154.2 ± 15.4*104.6 ± 16.1**73.1 ± 7.9**13.9 ± 13.3***0.000
72100136.7 ± 8.6*63.9 ± 13.6***44.1 ± 5.8***0***0.000
10% FBS
24186.9 ± 40.9186.1 ± 46.1141.3 ± 60.0128.7 ± 34.55.7 ± 0.9***0.007
48154.0 ± 44.5158.1 ± 43.4139.2 ± 20.0134.4 ± 28.512.4 ± 7.3***0.003
72151.5 ± 32.1159.6 ± 44.5125.6 ± 17.786.8 ± 39.99.0 ± 6.0***0.004

Serum samples from patients and healthy subjects were added at 2%, 10%, and 20% concentrations in culture medium. Cell viability significantly decreased in samples with increasing concentrations of sera from jaundiced patients at 72 hours (Table 3), with viability decreasing by 19%, 18%, and 33% at 2%, 10%, and 20% concentrations, respectively. No significant effect was observed at the other time points, although there was a trend in the experiments performed at 48 hours. Moreover, no effect on cell viability was observed in the experiments performed with samples from patients who had normal bilirubin levels.

Table 3. Effect of Sera from Healthy Subjects and Jaundiced and Nonjaundiced Patients on Osteoblast Viability
Time[%]HealthyNo JaundiceJaundiceP Value
  • *

    P = 0.036 versus healthy subjects (HS); **P = 0.02 versus HS; ***P = 0.008 versus HS. P Value: Multiple groups comparation. Significant differences are shown in bold/italic. Viability of primary human osteoblasts of three different pools of sera from 10 donors with increasing concentrations (2%, 10%, and 20%), performed at 24, 48, and 72 hours. Cell viability is expressed as a percentage of absorbance versus nontreated cells (control). Results are expressed as mean (±SD). Culture experiments were performed in triplicate for each pool of human osteoblast cells and treatments.

24 hours2100 ± 093.5 ± 7.893.4 ± 5.80.492
10109.3 ± 11.3107.0 ± 11.8104.6 ± 7.80.983
20102.6 ± 16.296.5 ± 11.476.8 ± 13.70.857
48 hours2100 ± 092.9 ± 3.192.1 ± 9.90.394
1088.6 ± 11.389.4 ± 5.472.6 ± 18.30.289
2088.2 ± 24.271.4 ± 20.751.5 ± 17.90.551
72 hours2100 ± 093.8 ± 8.280.2 ± 7.0*0.042
10105.9 ± 1.393.8 ± 3.181.1 ± 11.1*0.043
2094.7 ± 14.191.9 ± 8.867.2 ± 11.1*0.009

Effect of Bilirubin Concentration and Samples from Patients and Healthy Subjects on Osteoblast Differentiation.

These experiments were performed with unconjugated bilirubin concentrations below 1000 μM, because this latter concentration has very detrimental effects on osteoblast viability, as reported in the previous experiments. Bilirubin significantly decreased the alkaline phosphatase activity in primary human osteoblasts, with a clear-cut dose effect, because at 72 hours, differentiation decreased significantly by 14% and 55% at 50 μM and 100 μM bilirubin, respectively. Moreover, this detrimental effect of bilirubin was already observed with bilirubin at 100 μM at all time points (Fig. 1A). The presence of 10% FBS in the culture media prevented the detrimental effects on osteoblast differentiation, although there was a nonsignificant trend in the differentiation decreases (Fig. 1B).

thumbnail image

Figure 1. Effect of bilirubin and sera from healthy subjects and jaundiced and nonjaundiced patients on osteoblast differentiation. The experiments were performed with bilirubin at 10, 50, and 100 μM, with sera from healthy subjects and nonjaundiced and jaundiced patients in different conditions: (A) nonstimulated (basal) and (B) supplemented with 10% FBS, (C) 2% serum, (D) 10% serum and (F) 20% serum, at 24, 48, 72, and 96 hours. Alkaline phosphatase activity indicates differentiation. All experiments were performed in triplicate, and the results are expressed as mean ± SD. The experiments were performed with osteoblasts taken from 10 patients for each experiment. *P < 0.05 versus basal and FBS 10% experiments performed with or without bilirubin.

Download figure to PowerPoint

The addition of serum from jaundiced patients to cell cultures was also associated with reduced osteoblast differentiation, a finding that was already observed at the lowest concentration (2%) (Fig. 1C), being more evident with 10% and 20% plasma in the cultured media (Fig. 1D,F, respectively). Osteoblast differentiation was significantly diminished in experiments performed with sera from nonjaundiced patients as well, effects which were more evident with increasing concentrations, particularly at 96 hours (Fig. 1C,D,F). Thus, at 72 and 96 hours, the decrease in osteoblast differentiation was 16% and 54% for samples (2% concentration) from nonjaundiced patients, and 46% and 69% for samples from jaundiced patients, respectively (P ≤ 0.024). Significant decreases in osteoblast differentiation were also observed with 10% and 20% sera concentration from jaundiced and nonjaundiced patients. The highest concentration (20%) decreased osteoblast differentiation by 47% and 62% in nonjaundiced patients and 44% and 67% in jaundiced patients at 72 and 96 hours, respectively (P ≤ 0.011).

Effect of Bilirubin Concentration and Samples from Patients and Healthy Subjects on Osteoblast Mineralization.

Osteoblast mineralization, as measured by the alazarin red staining method, was significantly reduced in the experiments performed with 50 μM unconjugated bilirubin at all time points (reduction of 55%, 57%, 33%, and 32% bone nodule formation at 7, 14, 21, and 28 days of treatment, respectively), a finding which was not observed when 10 μM bilirubin was used (Fig. 2A). Moreover, the experiments carried out with serum from healthy subjects and patients indicated that adding jaundiced serum to the culture resulted in a significant decrease of cell mineralization at all times, except at 7 days after treatment, whereas no differences with respect to healthy subjects were observed in the experiments performed with serum from nonjaundiced patients (Fig. 2B).

thumbnail image

Figure 2. Effect of bilirubin and sera from jaundiced and nonjaundiced patients on osteoblast mineralization. SAOS-2 were treated with bilirubin at 10 or 50 μM in DMEM 10% FBS, and with sera from healthy subjects and nonjaundiced and jaundiced patients with 10% sera for 7, 14, 21, and 28 days. The data are means ± SD of three independent experiments. Mineralization is expressed as a mineral quantification (versus total protein). All experiments were performed in triplicate. *P < 0.001 with respect to osteoblasts cultured without bilirubin or healthy subjects.

Download figure to PowerPoint

Effect of Bilirubin Concentration and Samples from Patients and Healthy Subjects on mRNA Expression in Primary Human Osteoblasts.

Neither bilirubin nor jaundiced serum added to the osteoblast culture were associated with changes in the osteocalcin mRNA levels, although high concentrations of serum (20%) from patients and controls resulted in a decreased expression of osteocalcin mRNA.

Unconjugated bilirubin (50 μM) increased the expression of OPG and RANKL, effects which were more prominent with a higher concentration of FBS in the culture media. In contrast, no effects on these genes or in the RANKL/OPG ratio were observed with 10 μM bilirubin (Table 4). Moreover, bilirubin at 50 μM concentration significantly decreased the expression of RUNX2 (Table 4; Fig. 3A). The results observed in the experiments performed with serum from healthy subjects and patients were somewhat dissimilar, because expression of OPG decreased but RANKL increased, leading to a significant enhancement in the RANKL/OPG ratio (Table 4; Fig. 3B). In addition, no significant changes on RUNX2 expression were observed with pooled samples from patients and controls, although lower levels of mRNA expression were observed in parallel with increasing concentrations of serum (from 2% to 20%) in the culture media (Table 4).

Table 4. Influence of Bilirubin and Sera from Patients and Controls on Osteoprotegerin (OPG), RANKL, Osteocalcin (BGLAP), and RUNX2 Gene Expression
  BilirubinSamples from Patients and Healthy Subjects
  0 μM10 μM50 μMP Value[%]HealthyNo JaundiceJaundiceP Value
  • *

    versus 0 μM;

  • **

    versus 10 μM;

  • ***

    versus 10 and 50 μM.

  • * versus 2%; ** versus 5%;*** versus 2% and 5%

  • *versus healthy subjects (HS); **versus no jaundice (noJ); ***versus HS and NoJ

  • * versus HS; **versus noJ; ***versus HS and NoJ

  • P Value: Multiple groups comparation. Significant differences are shown in bold/italic. Data are expressed as relative units with respect to β-actin mRNA. n.s., not significant.

OPGFBS 2%1.36 ± 0.591.21± 0.351.63 ± 0.30n.s21.31 ± 0.141.17 ± 0.510.98 ± 0.12n.s.
 FBS 10%0.55 ± 0.131.12 ± 0.442.92 ± 0.30***<0.0001101.43 ± 0.471.53 ± 0.271.09 ± 0.43n.s.
      202.04 ± 0.541.37 ± 0.321.27 ± 0.38*0.026
RANKLFBS 2%1.35 ± 0.241.65 ± 0.622.96 ± 1.01*0.04921.05 ± 0.251.19 ± 0.390.48 ± 0.11**0.017
 FBS 10%1.26 ± 0.452.84 ± 0.933.99 ± 0.980.099101.61 ± 0.372.71 ± 0.632.72 ± 1.06n.s.
      201.33 ± 0.052.23 ± 0.474.57 ± 0.55<0.0001
RANKL/OPGFBS 2%1.32 ± 0.401.25 ± 0.271.78 ± 0.29n.s.20.83 ± 0.160.92 ± 0.360.50 ± 0.15***n.s.
 FBS 10%2.31 ± 0.042.69 ± 0.921.46 ± 0.24n.s.101.36 ± 0.431.54 ± 0.312.34 ± 0.19***0.009
      200.68 ± 0.131.66 ± 0.573.74 ± 1.00***0.005
BGLAPFBS 2%0.77 ± 0.490.72 ± 0.760.76 ± 0.55n.s.21.22 ± 0.121.15 ± 0.220.81 ± 0.240.066
 FBS10%0.33 ± 0.230.48 ± 0.300.52 ± 0.45n.s.100.55 ± 0.260.54 ± 0.130.45 ± 0.08n.s.
      200.38 ± 0.240.42 ± 0.160.41 ± 0.22n.s.
RUNX2FBS 2%0.61 ± 0.110.47 ± 0.080.23 ± 0.06*0.01521.02 ± 0.151.10 ± 0.320.95 ± 0.39n.s.
 FBS 10%1.58 ± 0.070.44 ± 0.210.42 ± 0.14***0.001100.46 ± 0.080.51 ± 0.210.62 ± 0.23n.s.
      200.33 ± 0.090.35 ± 0.020.49 ± 0.25n.s.
thumbnail image

Figure 3. Effect of bilirubin on (A) RUNX2 or (B) RANKL/OPG mRNA expression on human primary osteoblastic cells. Cells were treated with bilirubin at 10 or 50 μM in DMEM 10% FBS, and with sera from healthy subjects or nonjaundiced and jaundiced patients. The data are the means (± SD) of three to four independent experiments measured by real-time PCR. Gene expression is expressed as arbitrary units (versus β-actin). *P < 0.001 with respect to nontreated cells.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

The results of the current study, carried out using primary human osteoblasts, indicate that bilirubin has detrimental effects on cell viability, but also on osteoblast differentiation and mineralization. Thus, the presence of 50 μM unconjugated bilirubin in the culture media resulted in a decreased cell differentiation, as assessed by the alkaline phosphatase assay. Moreover, this concentration of bilirubin in the culture media decreased cell mineralization in SAOS-2 cells as well. The detrimental effects of bilirubin are in accordance with those induced by sera samples from jaundiced patients, even though the increased bilirubin was conjugated in these patients and the potential effects of other retained substances cannot be ruled out. In these experiments, 66 μM bilirubin, which was obtained in the experiments with 20% concentration, also decreased cell differentiation and mineralization, using similar approaches.

This study confirms previous data on the harmful effect of bilirubin on cell survival. Thus, as observed by Janes et al., the presence of sera with a high concentration of bilirubin resulted in decreased cell viability.5 Moreover, depression of proliferation of other cells of calcifying tissues by bilirubin has also been reported.21 The current study, however, adds new information, because reduced osteoblast viability was observed with sera samples from jaundiced patients, particularly in the experiments performed with the highest bilirubin concentration in culture media (66 μM). Conversely, serum from nonjaundiced patients had no detrimental effects or had lesser effects on survival. The differences in cell viability observed between the experiments performed with bilirubin in the media or with sera samples from healthy subjects and jaundiced and nonjaundiced patients may be explained, at least in part, by the presence of molecules other than bilirubin in the experiments. Among these other molecules, increased bile acid or different cytokines and growth factors released as a consequence of the pathological condition could participate in these detrimental effects on osteoblast function. Another result was the somewhat increased osteoblast viability observed in the experiments performed with physiological concentrations of unconjugated bilirubin, a phenomenon that has also been reported in other studies22 and which could be related to some antioxidant properties on unconjugated bilirubin.23-25

On the other hand, the findings from this study also indicate that unconjugated bilirubin and sera from jaundiced patients induced significant changes on gene expression, even in osteoblasts treated for a short time (that is, 24 hours). Thus, exposure to bilirubin at 50 μM decreased cell differentiation, as demonstrated by a significant down-regulation of RUNX2, effects which were already observed at low concentrations of sera (2%). This result is in agreement with the decreased alkaline phosphatase activity. The lack of effect of sera from jaundiced patients on RUNX2 gene expression can be, to some degree, explained by various factors: (1) increased total bilirubin and lesser amount of unconjugated bilirubin; (2) the presence of molecules other than bilirubin in the media; and (3) the short time used in the experiments, taking into account that osteocalcin is a late marker for osteoblast differentiation. These concerns can also account for the lack of effect of bilirubin and sera from jaundiced patients on osteocalcin gene expression, which was not significantly decreased in the experiments with primary human osteoblasts. This result is in contrast with other experiments performed in primary rat osteoblasts, indicating that very low bilirubin concentrations (3 and 30 μM) down-regulated the expression of osteocalcin and RUNX2 in osteoblasts cultured in osteogenic medium for 3 to 14 days.22

The OPG/RANKL system is the key regulator of osteoblast-induced osteoclastogenesis, because both osteoprotegerin and RANKL are primarily synthesized in osteoblasts but eventually act on osteoclasts, the cell line responsible for bone resorption. Thus, osteoblasts produce RANKL, a ligand for the receptor activator of nuclear factor-κB (RANK) on hematopoietic cells, which activates the differentiation of osteoclasts and maintains their function. Osteoblasts also produce and secrete osteoprotegerin, a decoy receptor that can block RANKL/RANK interactions.26 Even though the experiments were performed with a short period of incubation, jaundiced sera influenced the expression of RANKL and OPG. Serum from jaundiced patients was able to up-regulate RANKL and down-regulate OPG gene expression, thus resulting in a very significant increase in the RANKL/OPG gene expression ratio.

These novel data derived from this study can be extrapolated to what is observed in clinical conditions in patients with chronic cholestatic diseases, in whom, besides decreased bone formation, a parallel increase in bone resorption has been described,4, 27 an event that has been attributed to secondary hyperparathyroidism resulting from overt or subtle deficiencies in calcium and vitamin D levels as a consequence of intestinal malabsorption.1, 2, 27 Again, the lack of even a contradictory effect of unconjugated bilirubin alone by up-regulating OPG can be explained by the presence of other confounding molecules, such as increased bile acids or inflammatory cytokines, and growth factors in the sera from patients with cholestasis. It has been certainly demonstrated that lithocholic acid decreases the stimulatory effect of vitamin D on osteocalcin and RANKL mRNA expression in primary human osteoblasts,6 results which may explain the potential deleterious effects of retained bile acids on bone formation in patients with chronic cholestasis.

There are controversies on the potential detrimental effects of increased unconjugated hyperbilirubinemia (usually below 68 μM) associated with Gilbert's syndrome in the development of low bone mass and osteoporosis. Thus, some studies have pointed out an association between osteoporosis and Gilbert's syndrome,12 whereas other studies categorically oppose such association.11 A recent study of hyperbilirubinemic Gunn rats also failed to show a difference between bone mineral density and osteocalcin levels in hyperbilirubinenic rats and wild-type rats, suggesting that elevated serum bilirubin alone is not a major contributory factor to hepatic osteodystrophy that results in low bone mass in this animal model.11 The results of the current study cannot refute this latter study, but are in favor of such a relationship between low bone mass and hyperbilirubinemia, because the effect of unconjugated bilirubin on cell viability, differentiation, and mineralization were observed with bilirubin concentrations present in patients with Gilbert's syndrome (50 μM), but not with the experiments performed with normal bilirubin concentrations in nonjaundiced patients and in healthy subjects. The recent study showing a significant inverse correlation between unconjugated bilirubin levels and total body bone mineral density in a series of 17 subjects with Gilbert's syndrome may support our findings.12

In summary, our results indicate that unconjugated bilirubin and sera from jaundiced patients lead to clearly defective consequences in primary human osteoblasts and in an osteoblast cell line, besides decreased cell viability. Moreover, sera from jaundiced patients induced significant up-regulation of the RANKL/OPG ratio involved in osteoclastogenesis, and bilirubin down-regulated RUNX2 gene expression, a transcription factor related to osteoblastogenesis. Taken together, these data sustain the deleterious consequences of increased bilirubin in advanced chronic cholestatic diseases and in end-stage liver diseases on the development of bone loss, resulting from disturbed bone formation related to osteoblast dysfunction.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

This study was supported in part by CIBERehd and PI080105, Instituto de Salud Carlos III, Ministerio de Ciencia e Innovación, Spain.

References

  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References